The objectives of screening and diagnostic strategies in cancer are to detect disease within the window of curability and avoid over- and under-treatment. The discovery of imaging biomarkers that improve the identification of aggressive versus non-aggressive variants of disease would enhance our ability to treat patients appropriately. In contrast to normal cells, cancer cells typically reduce pyruvate to lactate, a phenomenon known as the Warburg effect. This altered metabolism provides an opportunity to distinguish tumors from healthy tissue by non-invasive molecular imaging. Increased demand for glucose by metabolically reprogrammed cancer cells is exploited by positron emission tomography (PET) imaging using [18F]fluorodeoxyglucose, [18F]FDG, which typically delineates tumors with high sensitivity. Prostate cancer (PCa), however, does not typically exhibit the Warburg effect, and [18F]FDG uptake in PCa is low. Alternative biomarkers to [18F]FDG would be valuable tools with which to diagnose and stage PCa, the most frequently diagnosed cancer in men in the US. As PCa develops resistance to therapies that block androgen receptor (AR) signaling, it transitions to a glycolytic phenotype characterized by reduction of pyruvate to lactate. Pyruvate is at the nexus of cellular metabolism, and is therefore an attractive candidate for imaging cancer metabolism by PET. Carbon-11-labeled pyruvate couples the advantages of PET, including enhanced sensitivity and biochemical resolution, to an essential metabolite whose specific metabolic fate has important implications for disease staging and prognosis, monitoring therapy, and identifying new therapeutic targets. We propose to label pyruvate with carbon-11 at either the C-1 position ([1- 11C]pyruvate) or the C-3 position ([3-11C]pyruvate). By labeling this simple molecule at different positions, it will be possible to infer the metabolic fate of pyruvate and therefore determine the aggressiveness of PCa tumors. Current syntheses of these molecules are inadequate for imaging needs. The automated synthesis of [1-11C]- and [3-11C]pyruvate via D-[11C]alanine is the first aim of this proposal. We will evaluate [1-11C]- and [3- 11C]pyruvate flux in two preclinical xenograft models of PCa: androgen-sensitive LNCaP, which we expect to predominantly oxidize pyruvate to acetyl-CoA and CO2, and androgen-insensitive, N-Myc overexpressing PC3, which we expect to predominantly reduce pyruvate to lactate. We hypothesize that the clearance of radioactivity from LNCaP tumors will be significantly faster following administration of [1-11C]pyruvate than following administration of [3-11C]pyruvate. By contrast, we expect the clearance rates to be similar in PC3 tumors. Finally, we will exploit known metabolic differences in PCa cell lines to measure the metabolic fate of pyruvate in the presence or absence of androgens. Our expectation is that aggressiveness and AR signaling blockade will be associated with elevated levels of phospho-Akt, N-Myc, and LDHA, and consequently increased reduction of pyruvate to lactate. On this basis, we aim to validate [1-11C]- and [3-11C]pyruvate as imaging biomarkers of cancer metabolism and tools for non-invasive assessment of tumor aggressiveness.